1 dzt
Jet Propulsion LaboratoryCalifornia Institute of Technology
Copyright2017. All rights reserved.
Type-II Superlattice (T2SL) Barrier Infrared Detector (BIRD)
for Earth Science Applications
David Z. Ting, Alexander Soibel, Arezou Khoshakhlagh, Sam A. Keo, Sir B. Rafol, Cory J. Hill, Anita M. Fisher, Edward M. Luong,
Brian J. Pepper, Jason M. Mumolo, John K. Liu, Sarath D. Gunapala
Center for Infrared Photodetectors NASA Jet Propulsion Laboratory
California Institute of Technology
NASA Earth Science Technology Forum 2017 (ESTF2017) June 14, 2017 - Beckman Institute, Caltech. Pasadena, California
2 dzt
Jet Propulsion LaboratoryCalifornia Institute of Technology
Copyright2017. All rights reserved.
Outline
• Introduction • Type-II Superlattice (T2SL) Barrier Infrared
Detector (BIRD)
• Summary
3 dzt
Jet Propulsion LaboratoryCalifornia Institute of Technology
Copyright2017. All rights reserved.
Introduction
4 dzt
Jet Propulsion LaboratoryCalifornia Institute of Technology
Copyright2017. All rights reserved.
Thermal Infrared Focal Plane Arrays for Earth Science
• Thermal infrared focal plane arrays (FPAs) for a variety of Earth Science related applications – Geology, ocean and ice changes, de-forestation, forest fires, soil moisture and
plant health, weather, gas detection, pollution monitoring, …
• Infrared band of interests – 3 – 5 μm MWIR atmospheric transmission window – 8 – 14 μm LWIR atmospheric transmission window – Outside transmission windows, e.g., λcutoff ~ 15.4 μm for atmospheric sounding
• Focal plane arrays needed for – Imaging – Spectral imaging (more demanding)
• Desired infrared FPA properties – Customizable cutoff wavelength – High operability, spatial uniformity, temporal stability, scalability, and affordability – Low dark current and high QE – Higher operating temperature, less demanding cooler – Reduced mass, volume, power
5 dzt
Jet Propulsion LaboratoryCalifornia Institute of Technology
Copyright2017. All rights reserved.
Traditional Bulk Infrared Material Cutoff Wavelength Coverage
• MCT is the most successful infrared material to date – Adjustable band gap covering NIR to VLWIR. Long τSRH. – Soft and brittle. Requires expert handling in growth, fabrication, storage. – Longer λcutoff, high Hg fraction, progressively more challenging
• FPAs based on (near) lattice-matched bulk III-V semiconductors are highly successful in a few cases – SWIR InGaAs on InP performs at near theoretical limit
• Single color, limited cutoff wavelength adjustability – InSb dominates MWIR market, despite lower operating temperature
• Fixed cutoff wavelength, single color – Lacking the continuous cutoff wavelength adjustability of MCT
2 4 6 8 10 12 14 16 [ µm ] CutoffWavelengthλc
II-VISemiconductors
BulkIII-VSemiconductors
HgCdTe(MCT)
InSb InGaAs
AtmosphericTransm.Windows LWIR/VLWIRMWIRSWIRNIR
6 dzt
Jet Propulsion LaboratoryCalifornia Institute of Technology
Copyright2017. All rights reserved.
Semiconductor IR Material on Available Substrates
• MCT grown on CZT (CdZnTe) substrate covers full range of infrared • In0.53Ga0.47As grown on InP substrate has ~1.7 µm cutoff wavelength (covers SWIR) • InSb grown on InSb substrate has 5.2 µm cutoff wavelength (covers MWIR)
1.0 µm
1.5 µm
5 µm 10 µm
Visible
Infrared
Substrates: GaAs InP InSb InAs GaSb CZT*
7 dzt
Jet Propulsion LaboratoryCalifornia Institute of Technology
Copyright2017. All rights reserved.
• Quantum well infrared photodetectors (QWIPs)
• Multi-band QWIPs
• Quantum dots (QDIPs)
• Type-II superlattice (T2SL) Barrier IR Detector
Adjustable λcwith III-V Quantum Structures Development at JPL Center for Infrared Photodetectors
1Kx1K LWIR QWIP
1Kx1K LWIR T2SL
1Kx1K MW/LW Dualband QWIP
1Kx1K LWIR QDIP
1Kx1K MWIR T2SL
320x256 LWIR T2SL
640x512 LWIR QDIP
8 dzt
Jet Propulsion LaboratoryCalifornia Institute of Technology
Copyright2017. All rights reserved.
Quantum Well Infrared Photodetector (QWIP)
• III-V semiconductor FPA “-ility” advantages – High operability, uniformity, large-format capability, producibility, affordability – Temporal stability (low 1/f noise). No need for frequent system recalibration.
• QWIP FPAs successfully deployed in LandSat-8, HyTES • QWIP Challenges
– Requires more cooling to control thermal dark current. Higher generation-recombination (G-R) rate from fast LO phonon scattering.
– Low external QE. Needs light coupling structure for normal-incidence absorption. • Being addressed in R-QWIP by K. K. Choi - Resonator pixel concept.
h�
GaAs
AlGaAs
E
z
EC
E1
E2
‒ GaAs/AlGaAs QW ‒ InGaAs/InAlAs QW ‒ Adjust λc thru well width/barrier height ‒ Multiple quantum wells between emitter
and collector contacts. Gain < 1.
E
k||
Re-combination via optical phonon emission
hn
E2(k||)
E1(k||)
9 dzt
Jet Propulsion LaboratoryCalifornia Institute of Technology
Copyright2017. All rights reserved.
Type-II Superlattice Barrier Infrared Detector
10 dzt
Jet Propulsion LaboratoryCalifornia Institute of Technology
Copyright2017. All rights reserved.
Concept and Theoretical Foundation of Superlattice Infrared Detectors - Caltech Connection
• Originally proposed for HgTe/CdTe superlattice – Key advantages of superlattice for infrared detection pointed out in 1983 MCT SL paper
• Subsequent focus on superlattices based on the antimonides material system – Smith & Mailhiot 1987 paper considered the seminal work in T2SL infrared detectors
Smith, McGill & Schulman (Caltech) Appl Phys Lett (1983) Advantages of MCT superlattices (SL IR detectors in general)
Schulman & McGill (Caltech) Appl Phys Lett (1979) MCT superlattice IR detector
Smith & Mailhiot (LANL & Xerox) J Appl Phys (1987) InAs/GaInSb strained type-II superlattices IR detector
G. C. Osbourn (Sandia) J Appl Phys (1982) Strained layer superlattice from lattice matched materials
G. C. Osbourn (Sandia) J Vac Sci Tech B (1984) InAsSb Strained layer superlattice for LWIR detector
Darryl Smith: Caltech thesis advisor of Gordon Osbourn Tom McGill: Caltech thesis advisor of Joel Schulman & Christian Mailhiot
11 dzt
Jet Propulsion LaboratoryCalifornia Institute of Technology
Copyright2017. All rights reserved.
Antimonides Material System for Type-II Superlattices
• Material system includes InAs, GaSb, AlSb and their alloys – Nearly lattice matched (~6.1 Å)
• Alloys with GaAs, AlAs, and InSb adds even more flexibility
• GaSb (2”,3”,4”, …) and InAs substrates
• Three types of band alignments – Type-I (nested, straddling) – Type-II staggered – Type-II broken gap (misaligned, Type-III)
• Unique among common semiconductor families
• Overlap between InAs CB and GaSb VB enables interband devices
• Tremendous flexibility in artificially designed materials / device structures
– Arsenides – Antimonides – Arsenide-Antimonides
12 dzt
Jet Propulsion LaboratoryCalifornia Institute of Technology
Copyright2017. All rights reserved.
GaSb Substrate Development
• GaSb substrate available commercially in 2, 3, and 4-inch formats – Cost:
13 dzt
Jet Propulsion LaboratoryCalifornia Institute of Technology
Copyright2017. All rights reserved.
Antimonide Type-II Superlattices
• Band gap can be made smaller than constituent bulk semiconductors • Continuously adjustable band gap / λcutoff by varying layer widths
– Covering SWIR, MWIR, LWIR, and VLWIR • Sufficiently large absorption coefficient to achieve ample QE • Dark current reduction in superlattice
– Can be engineered for Auger suppression – Less susceptible to tunneling reduction
• III-V semiconductor challenges – Generation-recombination (G-R) dark current due to SRH processes – Surface leakage dark current without good passivation
Adjustable λcutoff
EC
EV
InAsGa(In)Sb
Heavy-hole miniband
Conduction miniband
Light-hole miniband
Review Book Chapter: “Type-II Superlattice Infrared Detectors”, D. Z. Ting, A. Soibel, L. Höglund, J. Nguyen, C. J. Hill, A. Khoshakhlagh, and S. D. Gunapala, Semiconductors and Semimetals 84, pp.1-57 (2011).
14 dzt
Jet Propulsion LaboratoryCalifornia Institute of Technology
Copyright2017. All rights reserved.
Unipolar Barrier Detector Architecture: Addressing III-V Challenges
• Maimon & Wicks “nBn detector, an infrared detector with reduced dark current and higher operating temperature”, Appl Phys Lett. (2006) – 240 citations on Web of Science as of June 2017
• Arguably the most influential paper in infrared detectors in the past decade – The nBn and, in general, unipolar barrier infrared detectors (XBn,
pBp, DH, CBIRD, …) have been implemented in a wide variety of materials systems by research groups world-wide.
• The unipolar barrier in nBn blocks electrons but not holes – Leads to G-R and surface leakage dark current suppression
Contact
Unipolar electron barrier
n-type Absorber
B
n n
15 dzt
Jet Propulsion LaboratoryCalifornia Institute of Technology
Copyright2017. All rights reserved.
G-R Dark Current Suppression in nBn
• Conventional p-n diode – Defects in the band gap leads to SRH processes and G-R dark current in
depletion region – In many cases (e.g., InAs), surface of p-type layer inverts to n-type,
leading to surface leakage current path • The nBn
– SRH processes are drastically reduced in wide-band-gap barrier region – Suppresses G-R dark current – Photocurrent flows un-impeded – Barrier also blocks electron surface leakage current – Resulting in higher operating temperature / sensitivity
Eco
Ev n
Top Contact
Absorber
Electron Barrier
Thermal Thermal SRH
n B
Optical
Eco
Ev
p
n
Space charge region
SRH
Thermal
Thermal
Quasi-Neutral region Quasi-
Neutral region
D. Z. Ting 2008
Energy
Distance
nBn p-n diode
16 dzt
Jet Propulsion LaboratoryCalifornia Institute of Technology
Copyright2017. All rights reserved.
Example of T2SL based Unipolar Barrier Detector: Complementary Barrier Infrared Detector (CBIRD)
• Complementary Barrier Infrared Detector (CBIRD)
• p-type LWIR superlattice absorber
• unipolar hole barrier (hB) – widely adopted
• unipolar electron barrier (eB) • Both barriers are superlattice-
based
• Electron and hole barrier functions – Careful control of doping profile and placement of electrical (N-P) junction
inside hB suppresses G-R dark current without disrupting the extraction of minority carriers
– eB suppresses minority carrier injection (exclusion) – eB serves as a BSF layer; also suppresses electron surface leakage
Ting et al., Appl. Phys. Lett. 95, 023508 (2009); 102, 121109 (2013)
17 dzt
Jet Propulsion LaboratoryCalifornia Institute of Technology
Copyright2017. All rights reserved.
CBIRD Device Characteristics
• Discrete 200 µm × 200 µm devices • 9.8 mm cutoff (50% peak QE) • QE=40% (λ=8.5 µm, no AR coating) • Zero-bias turn-on • Jd( 0.1V, 77K) = 0.8x10-5 A/cm2 • Near-diffusion-limited dark current behavior to below 77K
Additional studies: • Gain and noise: Soibel et al., Appl. Phys. Lett. 96, 111102
(2010) • Proton radiation effect: Soibel et al., Appl. Phys. Lett. 107,
261102 (2015)
!ISC 0903 DI, 320x256, 30 mm pitch NEDT – 18.6 mK (f/2, 300K) [ Rafol et al., JQE 48, 878 (2012) ]
Ting et al., Appl. Phys. Lett. 102, 121109 (2013)
No A/R coating
18 dzt
Jet Propulsion LaboratoryCalifornia Institute of Technology
Copyright2017. All rights reserved.
JPL T2SL Barrier Infrared Detector (BIRD) FPAs
• Successfully implemented FPAs with a variety of λcutoff and formats • High operability/uniformity routinely achieved, MWIR to VLWIR.
λc ~ 5.4 µm
140 K
99.8% operability (320 x 256)
λc ~ 11 µm 78 K
λc ~ 11 µm
78 K
99.4% operability (1280 x 720 format)
99.1% operability (640 x 512)
99.94% / 99.95% operability (320 x 256 switchable)
λc ~ 12.4 µm
λc ~ 5.1 µm
61 K
Ting et al., Proc SPIE 10177, 101770N (2017)
19 dzt
Jet Propulsion LaboratoryCalifornia Institute of Technology
Copyright2017. All rights reserved.
T2SL BIRD FPA Development for Earth Science Applications
• CubeSat Infrared Atmospheric Sounder (CIRAS) – High Operating Temperature (HOT) BIRD
MWIR (λcutoff ~5.3 µm) FPA
• Hyperspectral Thermal Emission Spectrometer (HyTES) – LWIR (λcutoff ~12 µm) BIRD FPA to replace existing QWIP FPA – Higher QE, lower dark current, higher operating temperature – Retaining uniformity, operability, temporal stability …
• SLI-T: Long Wavelength Infrared FPA for Land Imaging – VLWIR (λcutoff ~13 µm) BIRD FPA – Goal: Significantly higher operating temperature than QWIP FPA – Plans for demonstrating a small sensor core as well as a very
large format FPAs in collaboration with industry partner
20 dzt
Jet Propulsion LaboratoryCalifornia Institute of Technology
Copyright2017. All rights reserved.
Summary
• Recent advances in III-V semiconductor IR detectors – Type-II superlattice (and bulk alloy) provides continuously
adjustable cutoff wavelength from SWIR to VLWIR – Unipolar barrier device architecture enhances detector
performance
• MWIR to VLWIR type-II superlattice barrier infrared detector (BIRD) FPAs routinely achieve high operability and uniformity
• Meeting a variety of Earth Science infrared FPA needs